WitnessAI Documentation
List
Start Here
Start Here
Analytics
Analytics
Conversations
Conversations
Users
Users
Alerts
Alerts
Policy
Policy
Administration
Administration
Apps & Resources
Apps & Resources
Release Notes
Release Notes
Administrator Guide
Administrator Guide
Welcome to WitnessAI!
Note: docs.witness.ai has a new layout, now matching the layout of the WitnessAI Console navigation menu.
← The Start Here section includes the Supported Apps & LLMs page.
← Below Start Here, each section on the left sidebar documents the menu items in the WitnessAI Console, as seen in the screenshot below. For example: Home, Discovery, Analytics, etc.
Integrations is a top-level section lower in the sidebar, and contains Identity, Network, and Witness Anywhere integration pages.
WitnessAI is an enterprise platform that helps you observe, control, and protect how artificial intelligence (AI) is used across your organization. It gives you visibility into AI usage, guardrails against risky behavior, and protection for your data, your employees, and your customers across thousands of AI applications and large language models (LLMs), including ChatGPT, Anthropic Claude, and Google Gemini.
Use this documentation to sign in, learn the console, and configure the controls that fit your organization's acceptable-use and compliance requirements.
What’s New?
Spend Dashboard
See full details in the November 3, 2026 Release Notes
Click here for all Release Notes.
Our Mission
We believe AI will be as fundamental to modern work as the web has been for the past twenty years. Our mission is to give organizations the security and governance controls they need to adopt AI safely.
WitnessAI addresses the core challenges that come with the rapid adoption of generative AI and LLMs: observability, control, and protection. By integrating with your existing network, application, and security infrastructure, WitnessAI delivers an encrypted, cloud-based service that monitors and governs AI interactions so they stay aligned with your internal policies and external regulatory obligations.
To sign in, see Log in to WitnessAI below. Once you’re signed in, start with Home to learn the dashboard.
After you’re familiar with the Console, refer to the User Guide for full details on every feature, and step-by-step usage instructions.
How WitnessAI Works
WitnessAI operates at the network level. It connects to your network security devices, identity providers (IdPs), and application interfaces to monitor and control AI interactions, and it integrates with Security Information and Event Management (SIEM) systems for real-time analysis of security-relevant activity.
WitnessAI captures employee access to internal and external AI applications and creates audit logs of all activity, including user prompts and AI responses during conversations. This lets you:
- Understand AI usage: see which applications and models people use, and how.
- Classify activity: every interaction is classified for risk and intent.
- Enforce policy: apply identity-based and intent-based policy to enforce your acceptable-use rules, including fine-grained data and topic controls.
- Protect your organization: prevent prompt injection and protect sensitive data and proprietary information.
Note: Deployment models vary. Most monitoring is delivered through network integration, while remote and off-network devices can be covered with Witness Anywhere. Work with your account team to choose the right model for your environment.
Get Started
- New to the console? Sign in (see Log in to WitnessAI below), then learn the dashboard on Home.
- Want to understand how the pieces fit together? See Concepts.
- Want to know what's covered? See Supported Applications & LLMs.
- Ready to set controls? See Policies and Policy Creation.
Log in to WitnessAI
- Open your browser and go to your organization's WitnessAI web console.
- On the Sign in page, enter your Email address and select Continue with email.
- On the next screen, enter your password, or choose to have a sign-in code emailed to you.
- If this is your first sign-in, or you haven't signed in recently, you may be emailed a six-digit verification code. Enter it to finish signing in.
After you sign in, the Home page is the first screen you see.
Note: Your organization may use single sign-on (SSO) through your identity provider instead of a password. If you can't sign in, contact your administrator.